The objective of the following case reports was to assess whether mineralized bone replacement grafts (eg, xenografts and allografts) could be added to recombinant human bone morphogenetic protein-2/acellular collagen sponge (rhBMP-2/ACS) in an effective manner that would: (1) reduce the graft shrinkage observed when using rhBMP-2/ACS alone, (2) reduce the volume and dose of rhBMP-2 required, and (3) preserve the osteoinductivity that rhBMP- 2/ACS has shown when used alone. The primary outcome measures were histomorphometric analysis of vital bone production and analysis of serial computed tomographic scans to determine changes in bone graft density and stability. Over the 6-month course of this investigation, bone graft densities tended to increase (moreso with the xenograft than the allograft). The increased density in allograft cases was likely the result of both compression of the mineralized bone replacement graft and vital bone formation, seen histologically. Loss of volume was greater with the four-sponge dose than the two-sponge dose because of compression and resorption of the sponges. Vital bone formation in the allograft cases ranged from 36% to 53% but, because of the small sample size, it was not possible to determine any significant difference between the 5.6 mL (four-sponge) dose and the 2.8 mL (two-sponge) dose. Histology revealed robust new woven bone formation with only minimal traces of residual allograft, which appeared to have undergone accelerated remodeling or rhBMP-2-mediated resorption.